creator
Built by one researcher,
for all researchers.
CropForge was created by Saswat Sundar Rath, a student researcher at ICAR-IARI Jharkhand, as an answer to a recurring problem — existing crop simulation tools were either closed-source, configuration-heavy, or too opaque to extend meaningfully.
The goal: a runtime that treats simulation as code, not configuration. Where every model, every parameter, and every assumption is explicit and inspectable. Where extending the physics means writing a Python class, not editing XML.
CropForge began as a concept — what if we could see the farm truly, in real time, on any device, with full freedom? A degree of visualisation once thought possible only on supercomputers. That game-like ambition became the stepping stone. Over successive releases it grew from a simple time-stepped prototype into a full-stack simulation runtime with opt-in physics, a terrain engine, sediment dynamics, WebGL visualisation, and first-class GLTF export.
// MIT licensed and built in the open. Contributions, critique, and collaboration are welcome.
How it grew.
v1.0.0 — Community Release + Weed competition, planting density, yield summary + CITATION.cff, Zenodo archival, JOSS-ready + Community governance, issue templates v0.9.x — Visual Architecture + PBR rendering, morph targets, machinery animation + Terrain-aware GLB export; stress/disease visualisation v0.8.0 — Terrain Arc + Sediment dynamics, mass conservation, D8 hydrology + LOD 3D renderer at 60 fps on 500×500 fields + TiedRidges, VegetativeFilterStrip, ContourBund v0.6–0.7 — Physics Arc + FAO-56 ET₀, Beer-Lambert radiation, RUSLE erosion + Terrain engine (procedural, CSV, GeoTIFF) + D8 hydrology coupling, 3D dashboard modal v0.1–0.5 — Core Runtime + Phase-ordered @step decorator, time-stepped simulation + Plugin architecture: StandardWheat, StandardMaize + Parquet logging, spatial state management